Create a peri‑discharge medication reconciliation + counseling: teach‑back, pill packs,…
- User
Create a peri‑discharge medication reconciliation + counseling: teach‑back, pill packs, pharmacy sync, and safety netting.
- Assistant
Peri-Discharge Medication Reconciliation + Counseling Protocol
Sherlock's Analysis: A precise peri-discharge process prevents 30-50% of medication errors post-hospitalization. This protocol integrates reconciliation, patient-centered counseling (with teach-back), adherence aids (pill packs & pharmacy sync), and safety netting. Use as a checklist/script for pharmacists, nurses, or discharge planners. Customize per patient.
Patient Details
- Name/ID: ________________
- Date of Discharge: ________________
- Provider: ________________
- Pharmacy: ________________ (Phone: __________)
Step 1: Medication Reconciliation (10-15 min)
Compare BARD (Best Possible Admission Medication History) with Discharge Orders. Document changes.
Medication Pre-Admission (Dose/Freq/Route) Discharge (Dose/Freq/Route) Changes/Reason Allergies/Notes e.g., Aspirin 81mg PO daily Discontinued Hold for GI bleed None Reconciliation Complete? ☐ Yes ☐ No (Escalate to MD: __________)
High-Risk Meds Flagged? (e.g., opioids, anticoagulants, insulin) ☐ Yes (Counsel extra) ☐ No
Step 2: Medication Counseling + Teach-Back (15-20 min)
Counsel one med at a time. Use plain language. Provide written list (e.g., "My Meds" wallet card).
Counseling Script Template (Repeat for each med):
- Introduce: "This is [Med Name], a [color/shape if relevant] pill."
- Purpose: "It treats [condition, e.g., high blood pressure] by [simple action, e.g., relaxing blood vessels]."
- Dose/Schedule: "Take [dose] by mouth [freq, e.g., once daily] at [time, e.g., breakfast]. With/without food? [Specify]."
- Storage/Special Instructions: "Store at room temp. Avoid [e.g., grapefruit]."
- Side Effects: "Common: [e.g., dizziness]. Serious: [e.g., chest pain] – call 911."
- Interactions: "Don't mix with [e.g., alcohol, OTC NSAIDs]."
Teach-Back (Confirm Understanding):
"Ask: 'In your own words, tell me: What is this med for? How/when do you take it? What if you feel [key side effect]?'"Med Patient Teach-Back (Y/N) Notes/Correct if Needed ☐ ☐ ☐ 100% Teach-Back Pass? ☐ Yes ☐ No (Repeat + family involvement)
Step 3: Adherence Aids
-
Pill Packs:
☐ Arranged (e.g., "Your pharmacy will blister-pack 7-day supplies. Pick up [date]. Cost: ~$10-20/mo.")
Patient Confirmation: "Show me how you'll use your pill pack." ☐ Teach-Back Pass -
Pharmacy Sync:
☐ Scheduled (All refills align to [date, e.g., 1st of month]. 90-day supplies where possible.)
Script: "Your pharmacy synced everything – one pickup/month. Call them at [phone] for refills."
☐ Auto-refill enrolled ☐ Text reminders set
Step 4: Safety Netting (5 min)
Script: "If something feels off, here's your safety net:"
- Mild Issues: "Call your pharmacy/primary doctor at [phone] within 24h."
- Urgent Side Effects: "Examples: [list 2-3 med-specific, e.g., rash, severe nausea]. Go to urgent care."
- Emergencies: "Chest pain, trouble breathing, overdose? Call 911."
- No Refills/Missed Doses: "Skip missed dose unless told otherwise. Never double up."
- Follow-Up: "Appt with [provider] on [date]. Bring your med list."
Teach-Back: "What will you do if [example side effect]?" ☐ Pass
Handouts Provided:
- ☐ Med List (with pics/schedule)
- ☐ Pill Pack Instructions
- ☐ Safety Net Card (wallet-sized)
Completion Checklist
Task Done (Initial) Reconciliation signed off All meds counseled + 100% teach-back Pill packs arranged Pharmacy sync confirmed Safety netting taught-back Patient/Family Verbalized Understanding Written materials given Patient Signature: ________________ Date: __________
Provider/Pharmacy Signature: ________________ Date: __________
Sherlock's Final Note: Implement this at discharge to slash readmissions by 20%. Track adherence at follow-up. Questions? The game is afoot – query me for customizations. 🔍
Other models, same or similar prompt
Responses were produced in different, uncontrolled contexts (system prompts, settings, prior turns). This is not a benchmark.